• Journal of Welding and Joining
• /
• v.16 no.5
• /
• pp.76-85
• /
• 1998

The resistance spot welding process has been used for joining the sheet metal in automotive engineering. In the resistance spot welding, the weld quality is very important, because the quality of weld is one of the most important factors to the automobile quality. The size of he molten nugget has been utilized to estimate the weld quality. However, it is not easy to find the weld defects. For weldability estimation, we have to use the nondestructive method such as X-ray or ultrasonic inspection. But these kinds of approaches are not suitable for detecting the defects in real time. The purpose of this study is to develop the real time monitoring of the weld quality in the resistance spot welding. Obtained data were used to estimate weldability using fuzzy algorithm. It is sound that this monitoring and estimation system can be useful to improve the weld quality in the resistance spot welding process and it is possible to estimate the weldability in real time.

• Journal of Ocean Engineering and Technology
• /
• v.11 no.1
• /
• pp.36-43
• /
• 1997

Weld bead shape is an important measure for evaluation of weld quality. Many welding parameters have influence on the weld bead shape. The quantitative relationship between welding parameters and bead shape, however, is not determined yet because of their high complexity and many unknown factors. Fuzzy expert system is an advanced expert system which uses fuzzy rules and approximate reasoning. It is a vert useful tool for welding technology because is can process rationally the uncertain and inexact information such as the welding information. In this paper, the empirical and the qualitative relationship between welding parameters and bead shape are analyzed and represented by fuzzy rules. They are converted to the quantitative relationship by use of approximate reasoning of fuzzy expert system. Weld bead shape is estimated from the welding parameters using fuzzy expert system. The result of comparison between measured values of weld bead by welding experiments and the estimates values by fuzzy expert system shows a good consistancy.

• Proceedings of the KWS Conference
• /
• 2000.04a
• /
• pp.303-306
• /
• 2000

Quality estimation of the weld has been one of the important issues in RSW which is a main process of the sheep metal fabrication in auto-body industry, It was well known that among the various welding process variables, dynamic resistance has a close relation with nugget formation. With this variable, it is possible to estimate the weld quality in real time. In this study, a new quality estimation algorithm is developed with the primary dynamic resistance measured at welding machine timer. For this, feature recognition method of Hopfield neural network is used. Primary resistance patterns are vectorized and classified with five patterns. The network trained by these patterns recognizes the dynamic resistance pattern and estimates the weld quality Because the process variable monitored at the primary circuit is used, it is possible to apply this system to real time application without any consideration of electrode wear or shunt effect.

• Proceedings of the KWS Conference
• /
• 1999.10a
• /
• pp.142-145
• /
• 1999

The dynamic resistance monitoring is one of the important issues in that in-process and real time quality assurance of resistance spot weld is needed to increase the product reliability. Secondary dynamic resistance patterns, as a real manner, are hard to adapt those factors in real time and in-plant system. In the present study, a new dynamic resistance detecting method is presented as a practical manner of weld quality assurance at the primary circuit. By the correlation analysis, it is found that the primary dynamic resistance patterns are basically similar to those of the secondary. Various dynamic resistance indices are characterized with the primary curve. And quality of the weld, like the tensile shear strength, is estimated using adaptive neuro-fuzzy estimation system which is consisted of the Sugeno fuzzy algorithm. Through the fuzzy clustering and parameter optimization, real time weld quality assurance system with less efforts is proposed.

• Proceedings of the KWS Conference
• /
• 2002.10a
• /
• pp.330-335
• /
• 2002

Resistance spot welding is one of the most widely used processes in sheet metal fabrication. Quality assurance of welding has been important to increase the productivity. In this study, weld quality estimation using primary circuit dynamic resistance applied to the in-process real-time systems. For quality estimation, factors relating to quality were extracted from the dynamic resistance, measured in the timer. The relationship between these factors and weld quality was determined through a artificial neural network model. This method has the advantage over the conventional one, such as obtaining the quality information without the use of extra devices.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.24 no.10 s.181
• /
• pp.2581-2588
• /
• 2000

In previous studies, the dynamic resistance, which was calculated by the process variables measured at the electrode of the welding machine, and the electrode displacement were used for quality exa mination. However, in-process usage of such systems is not effective in systems that include a welding gun attached to a robot. In order to overcome such problems, we obtained and used the process variables from the welding machine timer. This would allow us to estimate real time in -process weld quality. For quality estimation, the features were extracted as factors from the primary dynamic resistance patterns, which were measured in t he welding machine timer. The relationship between the indexes and nugget size of the welds was observed through the regression analysis. Using the analyzed factors, a regression model that could estimate nugget diameter was developed. Two regression equations of the model were suggested depending on the factors, and it was showed that the model developed by stepwise method was effective one for weld quality estimation. The developed estimation model was in good linearity with the nugget diameter obtained through the experimentation.

• Journal of the Korean Society of Manufacturing Technology Engineers
• /
• v.7 no.6
• /
• pp.57-63
• /
• 1998

An adaptive control in the robotic GMA welding is employed to monitor information about weld characteristics and process parameters as well as to modify those parameters to hold weld quality within acceptable limits. Typical characteristics are the bead geometry, composition, microstructure, appearance, and process parameters which govern the quality of the final weld. The main objectives of this thesis are to realize the mapping characteristics of bead width through learning. After learning, the neural estimation can estimate the bead width desired form the learning mapping characteristic. The design parameters of the neural network estimator(the number of hidden layers and the number of nodes in a layer) are chosen from an estimation error analysis. A series of bead of bead-on-plate GMA welding experiments was carried out in order to verify the performance of the neural network estimator. The experimental results show that the proposed neural network estimator can predict the bead width with reasonable accuracy and guarantee the uniform weld quality.

• Transactions of the Korean Society of Mechanical Engineers
• /
• v.17 no.2
• /
• pp.393-406
• /
• 1993

In resistance spot welding process, size of molten nuggest have been utilized to assess the integrity of the weld quality. However real-time monitoring of the nugget size is an extremely difficult problem. This paper describes the design of an artificial neural networks(ANN) estimator to predict the nugget size for on-line use of weld quality monitoring. The main task of the ANN estimator is to realize the mapping characteristics from the sampled dynamic resistance signal to the actual negget size through training. The structure of the ANN estimator including the number of hidden layers and nodes in a layer is determined by an estimation error analysis. A series of welding experiments are performed to assess the performance of the ANN estimator. The results are quite promissing in that real-time estimation of the invisible nugget size can be achieved by analyzing the dynamic resistance signal without any conventional destructive testing of welds.

• 제어로봇시스템학회:학술대회논문집
• /
• 1996.10b
• /
• pp.1404-1407
• /
• 1996

A quality monitoring system in butt welding process is proposed to estimate weld pool sizes. The geometrical parameters of the weld pool such as the top bead width and the penetration depth plus half back width are utilized to prove the integrity of the weld quality. The monitoring variables used are the surface temperatures measured at three points on the top surface of the weldment. The temperature profile is assumed that it has a gaussian distribution in vertical direction of torch movement and verify this assumption through temperature analysis. A neural network estimator is designed to estimate weld pool size from temperature informations. The experimental results show that the proposed neural network estimator which used gaussian distribution as temperature information can estimate the weld pool sizes accurately than used three point temperatures as temperature information. Considering the change of gap size in butt welding, the experiment were performed on various gap size.

• Proceedings of the KWS Conference
• /
• 2002.10a
• /
• pp.307-313
• /
• 2002

This paper describes a weld pool monitoring technique, which is based on the weld pool image analysis. The proposed image analysis algorithm uses machine vision techniques to extract geometrical information from the weld pool image such as maximum weld pool width, gap width and misalignment between the joint longitudinal axis and the welding wire. These can be related to the welding parameters (welding voltage and current, wire feed speed and standoff) to produce control actions necessary to ensure that the required weld quality will be achieved. The experiments have shown that the algorithm is able to produce good estimates of the weld pool geometry; however, the adjustment of the camera parameters affects the image quality and, consequently, has a great influence over the estimation.